- 08 Jan, 2018 12 commits
-
-
Daniel Jurgens authored
When mlx5_ib_add is called determine if the mlx5 core device being added is capable of dual port RoCE operation. If it is, determine whether it is a master device or a slave device using the num_vhca_ports and affiliate_nic_vport_criteria capabilities. If the device is a slave, attempt to find a master device to affiliate it with. Devices that can be affiliated will share a system image guid. If none are found place it on a list of unaffiliated ports. If a master is found bind the port to it by configuring the port affiliation in the NIC vport context. Similarly when mlx5_ib_remove is called determine the port type. If it's a slave port, unaffiliate it from the master device, otherwise just remove it from the unaffiliated port list. The IB device is registered as a multiport device, even if a 2nd port is not available for affiliation. When the 2nd port is affiliated later the GID cache must be refreshed in order to get the default GIDs for the 2nd port in the cache. Export roce_rescan_device to provide a mechanism to refresh the cache after a new port is bound. In a multiport configuration all IB object (QP, MR, PD, etc) related commands should flow through the master mlx5_core_dev, other commands must be sent to the slave port mlx5_core_mdev, an interface is provide to get the correct mdev for non IB object commands. Signed-off-by:
Daniel Jurgens <danielj@mellanox.com> Reviewed-by:
Parav Pandit <parav@mellanox.com> Signed-off-by:
Leon Romanovsky <leon@kernel.org> Signed-off-by:
Jason Gunthorpe <jgg@mellanox.com>
-
Daniel Jurgens authored
When multiple RoCE ports are supported registration for events on multiple netdevs is required. Refactor the event registration and handling to support multiple ports. Signed-off-by:
-
Daniel Jurgens authored
Remove use of the num_ports general capability throughout. The number of ports will be variable in the future, and reported in a different way. Signed-off-by:
-
Daniel Jurgens authored
It always returns 0. Change return type to void. Signed-off-by:
-
Daniel Jurgens authored
Generate a unique 128bit identifier for each host and pass that value to firmware in the INIT_HCA command if it reports the sw_owner_id capability. Each device bound to the mlx5_core driver will have the same software owner ID. In subsequent patches mlx5_core devices will be bound via a new VPort command so that they can operate together under a single InfiniBand device. Only devices that have the same software owner ID can be bound, to prevent traffic intended for one host arriving at another. The INIT_HCA command length was expanded by 128 bits. The command length is provided as an input FW commands. Older FW does not have a problem receiving this command in the new longer form. Signed-off-by:
-
Daniel Jurgens authored
There are two potential problems with the existing implementation. 1. Enable and disable can race after the atomic operations. 2. If a command fails the refcount is left in an inconsistent state. Introduce a lock and perform error checking. Fixes: a6f7d2af ("net/mlx5: Add support for multiple RoCE enable") Signed-off-by:
-
Moni Shoua authored
A DC Target (DCT) QP is represented in the hardware as a unique object. This object is created by CREATE_DCT command and destroyed by DESTROY_DCT command. However, in the driver we describe it as a QP. The hardware command that creates a DCT needs parameters that the verb create_qp() does not provide. Those remaining parameters are provided with the call to the verb modify_qp(). Therefore we delay the actual creation of a DCT in the hardware until the stage of modify_qp() to RTR. A support for query_qp() was added as well. It uses QUERY_DCT command to retrieve the applicable fields. Signed-off-by:
Moni Shoua <monis@mellanox.com> Reviewed-by:
Yishai Hadas <yishaih@mellanox.com> Signed-off-by:
-
Moni Shoua authored
DC Initiator (DCI) QP is represented like any other QP in the hardware. However, like any other transport QP there are attributes and settings that are special to DCI QP and needs specific attention and care. Make necessary changes to configure DCI QP. Signed-off-by:
-
Moni Shoua authored
The QP type IB_QPT_DRIVER doesn't describe the transport or the service that the QP provides but those are known only to the hardware driver. The actual type of the QP is stored in the hardware driver context (i.e. mlx5_qp) under the field qp_sub_type. Take the real QP type and any extra data that is required to create the QP from the driver channel and modify the QP initial attributes before continuing with create_qp(). Downstream patches from this series will add support for both DCI and DCT driver QPs. Signed-off-by:
-
Moni Shoua authored
Vendors can implement type of QPs that are not described in the InfiniBand specification. To still be able to use the IB/core layer services (e.g. user object management) without tainting this layer with driver proprietary logic, a new QP type is added - IB_QPT_DRIVER. This will be a general QP type that the core layer doesn't know about its true nature. When a command like create_qp() is passed to a hardware driver the extra data that is required is taken from the driver channel. Downstream patches from this series will use that QP type in the mlx5 driver. Signed-off-by:
-
Moni Shoua authored
Enable DC transport in the firmware to provide its functionality. Signed-off-by:
-
Moni Shoua authored
Add a missing command interface to work with a DCT. It includes: creating, destroying and get events for. Signed-off-by:
-
- 05 Jan, 2018 11 commits
-
-
Jonathan Toppins authored
Reporting that a device doesn't support RoCE seems like a valuable piece of information to have when trying to determine why a driver is not binding to a device. Better to report this at info log level instead of requiring a user to enable all debug messages in the driver. Signed-off-by:
Jonathan Toppins <jtoppins@redhat.com> Acked-By:
Devesh Sharma <devesh.sharma@broadcom.com> Signed-off-by:
Doug Ledford <dledford@redhat.com>
-
Hans Westgaard Ry authored
The cases for FDR/EDR signalling speed, were missing in ib_rate_to_mult and mult_to_ib_rate giving wrong return values when drivers convert static rate to/from inter-packet-delay. Signed-off-by:
Hans Westgaard Ry <hans.westgaard.ry@oracle.com> Reviewed-by:
Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by:
-
Jia-Ju Bai authored
i40iw_wait_pe_ready is not called in an interrupt handler nor holding a spinlock. The function mdelay in it can be replaced with msleep, to reduce busy wait. Signed-off-by:
Jia-Ju Bai <baijiaju1990@gmail.com> Reviewed-by:
Leon Romanovsky <leonro@mellanox.com> Acked-by:
Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by:
-
Kaike Wan authored
This patch adds static trace for RNRNAK timer. Currently the output from hrtimer static trace only shows the addresses of hrtimers in the system and there is no easy way to correlate an RNRNAK timer with its entries in the hrtimer trace. This patch adds the correlation among a QP, its RNRNAK timer, and its entries in the hrtimer trace. This correlation will be enormously helpful when debugging RNRNAK related issues. In addition, this patch cleans up rvt_stop_rnr_timer() to be void while here. Reviewed-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Reviewed-by:
Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by:
Kaike Wan <kaike.wan@intel.com> Signed-off-by:
-
Michael J. Ruhl authored
The get_unit_name() function crafts a string based on the device name and the device unit number. It then stores this in a static variable. This has concurrency issues as can be seen with this log: hfi1 0000:02:00.0: hfi1_1: read_idle_message: read idle message 0x203 hfi1 0000:01:00.0: hfi1_1: read_idle_message: read idle message 0x203 The PCI device ID (0000:02:00.0 vs. 0000:01:00.0) is correct for the message, but the device string hfi1_1 is incorrect (it should be hfi1_0 for the second log message). Remove get_unit_name() function. Instead, use the rvt accessor rvt_get_ibdev_name() to get the IB name string. Clean up any hfi1_early_xx calls that can now use the new path. QIB has the same (qib_get_unit_name()) issue. Updating as necessary. Remove qib_get_unit_name() function. Update log message that has redundant device name. Reviewed-by:
Michael J. Ruhl <michael.j.ruhl@intel.com> Signed-off-by:
-
Mike Marciniszyn authored
CQ allocation does not ensure that completion queue entries and the completion queue structure are allocated on the correct numa node. Fix by allocating the rvt_cq and kernel CQ entries on the device node, leaving the user CQ entries on the default local node. Also ensure CQ resizes use the correct allocator when extending a CQ. Reviewed-by:
Sebastian Sanchez <sebastian.sanchez@intel.com> Signed-off-by:
-
Sebastian Sanchez authored
When an 8051 command times out, the entire DC block is restarted. During the restart, the host interface version bit is set, which calls do_8051_command() recursively. The host version bit needs to be set before the link moves into polling, so the host version bit can be set in set_local_link_attributes() instead. Thus, the 8051 command functions can be simplied as a non-locking version (dd->dc8051_lock) of those functions are no longer needed. Fixes: 9be6a5d7 ("IB/hfi1: Prevent LNI out of sync by resetting host interface version") Reviewed-by:
-
Mike Marciniszyn authored
Normal receive queue allocation ensures that kernel receive queues are allocated on the local numa node. Shared receive queues do not behave the same way. Ensure that kernel shared receive queues are allocated on the device local node. Reviewed-by:
-
Michael J. Ruhl authored
rdmavt has a down call to client drivers to retrieve a crafted card name. This name should be the IB defined name. Rather than craft the name each time it is needed, simply retrieve the IB allocated name from the IB device. Update the function name to reflect its application. Clean up driver code to match this change. Reviewed-by:
-
Michael J. Ruhl authored
Currently the HFI and QIB drivers allow the IB core to assign a unit number to the driver name string. If multiple devices exist in a system, there is a possibility that the device unit number and the IB core number will be mismatched. Fix by using the driver defined unit number to generate the device name. Reviewed-by:
-
Kaike Wan authored
When the rdmavt's RNRNAK timer is fired, it tries to cancel the timer by calling hrtimer_try_to_cancel(), which always returns -1 because the timer is currently running. This patch removes this useless call. Reviewed-by:
-
- 04 Jan, 2018 7 commits
-
-
Bart Van Assche authored
Eliminate the WARN_ONs that create following two warnings when registering an rxe device: WARNING: CPU: 2 PID: 1005 at drivers/infiniband/core/device.c:449 ib_register_device+0x591/0x640 [ib_core] CPU: 2 PID: 1005 Comm: run_tests Not tainted 4.15.0-rc4-dbg+ #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/01/2014 RIP: 0010:ib_register_device+0x591/0x640 [ib_core] Call Trace: rxe_register_device+0x3c6/0x470 [rdma_rxe] rxe_add+0x543/0x5e0 [rdma_rxe] rxe_net_add+0x37/0xb0 [rdma_rxe] rxe_param_set_add+0x5a/0x120 [rdma_rxe] param_attr_store+0x5e/0xc0 module_attr_store+0x19/0x30 sysfs_kf_write+0x3d/0x50 kernfs_fop_write+0x116/0x1a0 __vfs_write+0x23/0x120 vfs_write+0xbe/0x1b0 SyS_write+0x44/0xa0 entry_SYSCALL_64_fastpath+0x23/0x9a WARNING: CPU: 2 PID: 1005 at drivers/infiniband/core/sysfs.c:1279 ib_device_register_sysfs+0x11d/0x160 [ib_core] CPU: 2 PID: 1005 Comm: run_tests Tainted: G W 4.15.0-rc4-dbg+ #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/01/2014 RIP: 0010:ib_device_register_sysfs+0x11d/0x160 [ib_core] Call Trace: ib_register_device+0x3f7/0x640 [ib_core] rxe_register_device+0x3c6/0x470 [rdma_rxe] rxe_add+0x543/0x5e0 [rdma_rxe] rxe_net_add+0x37/0xb0 [rdma_rxe] rxe_param_set_add+0x5a/0x120 [rdma_rxe] param_attr_store+0x5e/0xc0 module_attr_store+0x19/0x30 sysfs_kf_write+0x3d/0x50 kernfs_fop_write+0x116/0x1a0 __vfs_write+0x23/0x120 vfs_write+0xbe/0x1b0 SyS_write+0x44/0xa0 entry_SYSCALL_64_fastpath+0x23/0x9a The code should accept either a parent pointer or a fully specified DMA specification without producing warnings. Fixes: 99db9494 ("IB/core: Remove ib_device.dma_device") Signed-off-by:
Bart Van Assche <bart.vanassche@wdc.com> Cc: Leon Romanovsky <leon@kernel.org> Cc: stable@vger.kernel.org # v4.11 Signed-off-by:
-
Mark Bloch authored
Unconditional locks/list and ODP srcu initialization should be done in the INIT stage. Remove those from the CAPS stage and move them to the proper stage. Signed-off-by:
Mark Bloch <markb@mellanox.com> Signed-off-by:
-
Mark Bloch authored
The loopback stage only initializes a lock, move it to be in the CAPS initialization phase and get rid loopback step completely. Signed-off-by:
-
Mark Bloch authored
Now that we have a stage just for hardware counters, move all relevant initialization logic into one place. Signed-off-by:
-
Mark Bloch authored
Now that we have a stage just for ODP, move all relevant initialization logic into one place. Signed-off-by:
-
Mark Bloch authored
Now that we have a stage just for RoCE/ETH, move all relevant initialization logic into one place. Signed-off-by:
-
Mark Bloch authored
Today we have single function which is used when we add an IB interface, break this function into multiple functions. Create stages and a generic mechanism to execute each stage. This is in preparation for RDMA/IB representors which might not need all stages or will do things differently in some of the stages. This patch doesn't change any functionality. Signed-off-by:
-
- 03 Jan, 2018 8 commits
-
-
Michael J. Ruhl authored
When debugging issues with RC QPs, it is useful to know if a QP has an associated RQ or SRQ, the size of the RQ, and any RNR timeout values. Add the necessary information to the QP stats output. Reviewed-by:
-
oulijun authored
It needs to create eight reserve QPs for resolving a bug of hip06. When deregistering mr, it will issue a rdma write for every reserve QPs. When modify qp from init to rtr, it needs to set the value of dest_qp_num. Otherwise, it will lead an error of freeing mr. Signed-off-by:
Lijun Ou <oulijun@huawei.com> Signed-off-by:
-
Yixian Liu authored
The QP can accept send work requests only when the QP is in the states that allow them to be submitted. This patch updates the QP state judgement based on the specification. Signed-off-by:
Yixian Liu <liuyixian@huawei.com> Signed-off-by:
Shaobo Xu <xushaobo2@huawei.com> Signed-off-by:
-
oulijun authored
When the length of sge is zero, the driver need to filter it Signed-off-by:
-
oulijun authored
This patch refactors the code of setting access flags for RDMA operation as well as adds the scene when attr->max_dest_rd_atomic is zero. Signed-off-by:
-
oulijun authored
This patch fixes the usage with sr_max filed and rr_max of qp context when modify qp. Its modifications include: 1. Adjust location of filling sr_max filed of qpc 2. Only assign the number of responder resource if IB_QP_MAX_DEST_RD_ATOMIC bit is set 3. Only assign the number of outstanding resource if IB_QP_MAX_QP_RD_ATOMIC 4. Fix the assgin algorithms for the field of sr_max and rr_max of qp context Signed-off-by:
Wei Hu (Xavier) <xavier.huwei@huawei.com> Reviewed-by:
-
oulijun authored
This patch mainly implement rq inline data feature for hip08 RoCE in kernel mode. Signed-off-by:
-
Alex Estrin authored
If IB_CQ_REPORT_MISSED_EVENTS flag is passed in ib_req_notify_cq() it may return positive value indicating non-empty CQ. If return code not verified the log might be flooded with false warning messages "request notify on send CQ failed". Fixes: 8966e28d ("IB/ipoib: Use NAPI in UD/TX flows") Reviewed-by:
Alex Estrin <alex.estrin@intel.com> Signed-off-by:
-
- 02 Jan, 2018 2 commits
-
-
Leon Romanovsky authored
The commit 1a1c116f ("RDMA/netlink: Simplify the put_msg and put_attr") removes nlmsg_len calculation in ibnl_put_attr causing netlink messages and caused to miss source and destination addresses. Fixes: 1a1c116f ("RDMA/netlink: Simplify the put_msg and put_attr") Signed-off-by:
-
Leon Romanovsky authored
The NLDEV commands are using IB device indexes and names as a handle for netlink communications. Put all relevant code into one function to remove code duplication in followup patches. Reviewed-by:
-
